The Known Antimammalian and Insecticidal Alkaloids Are Not Responsible for the Antifungal Activity of Epichloë Endophytes

Asexual Epichloë sp. endophytes in association with pasture grasses produce agronomically important alkaloids (e.g., lolitrem B, epoxy-janthitrems, ergovaline, peramine, and lolines) that exhibit toxicity to grazing mammals and/or insect pests. Novel strains are primarily characterised for the presence of these compounds to ensure they are beneficial in an agronomical setting. Previous work identified endophyte strains that exhibit enhanced antifungal activity, which have the potential to improve pasture and turf quality as well as animal welfare through phytopathogen disease control. The contribution of endophyte-derived alkaloids to improving pasture and turf grass disease resistance has not been closely examined. To assess antifungal bioactivity, nine Epichloë related compounds, namely peramine hemisulfate, n-formylloline-d3, n-acetylloline hydrochloride, lolitrem B, janthitrem A, paxilline, terpendole E, terpendole C, and ergovaline, and four Claviceps purpurea ergot alkaloids, namely ergotamine, ergocornine, ergocryptine, and ergotaminine, were tested at concentrations higher than observed in planta in glasshouse and field settings using in vitro agar well diffusion assays against three common pasture and turf phytopathogens, namely Ceratobasidium sp., Drechslera sp., and Fusarium sp. Visual characterisation of bioactivity using pathogen growth area, mycelial density, and direction of growth indicated no inhibition of pathogen growth. This was confirmed by statistical analysis. The compounds responsible for antifungal bioactivity of Epichloë endophytes hence remain unknown and require further investigation.


Introduction
Asexual Epichloë sp. are endophytic fungi that play a critical role in pastoral agricultural systems as they enhance pasture performance. Improved growth in pastures is achieved in part from the bioprotective properties endophytes confer, including pest and disease resistance as well as abiotic stress tolerance [1][2][3][4]. While the impact of alkaloids on animal performance and insect control is well understood [5][6][7][8], the bioprotective compounds responsible for improved pasture quality and performance, in particular disease control, have not been thoroughly investigated.
Pasture and turf grasses are threatened by many phytopathogens, including fungi, bacteria, and viruses. Severe disease outbreaks impact pasture quality and yield [9][10][11]. Adding to the effect on the dairy and meat industries by reduced yields, mycotoxins produced by some phytopathogenic fungi are a health threat to grazing mammals [12][13][14]. Better understanding of Epichloë-mediated disease resistance, and identifying the compounds responsible, provides an opportunity to identify Epichloë endophyte strains better suited for pastoral agriculture and turf systems.
While the asexual (non-pathogenic) Epichloë sp. utilised in pasture and turf grasses have not been investigated with the same level of detail, bioactive antifungal compounds have been identified and isolated from sexual (pathogenic) Epichloë sp. Antifungal compounds that have been identified and isolated from Epichloë typhina include chokols (A-G), gamahonolide A, gamahonolide B, gamahorin, and epichlicin [24][25][26][27]. Several studies also describe isolation of antifungal agents from Epichloë festucae. An antifungal protein (55 amino acids, 6.3 kDa) named Efe-AfpA was isolated from E. festucae (Rose City isolate) and was shown to inhibit the growth of Sclerotinia homoeocarpa in a plate assay. Another antibiotic protein, VibA, was identified when the gene was overexpressed in E. festucae strain E437. The authors determined the VibA gene encoded for a biosynthetic enzyme that produced ε-poly-L-lysines of 28-34 lysine subunits. The antibiotic metabolites indole-3acetic acid, indole-3-ethanol, methylindole-3-carboxylate, indole-3-carboxaldehyde, N,Ndiacetamide, cyclonerodiol (an isomer of chokol C) have been isolated from E. festucae (BM7, M. D. Richardson) and showed inhibition against Cryphonectria parasitica, Magnaporthe poae, Laetisaria fuciformis, and Rhizoctonia solani [17,21,28,29]. Fernando et al. (2020) previously identified variation for antifungal activity between different strains of asexual Epichloë sp. isolated from perennial ryegrass (Lolium perenne) and tall fescue (Festuca arundinacea) plants [20,30,31]. The study showed that unique compounds produced by three highly bioactive endophyte strains, namely SE, NEA12, and NEA23, may be associated with the distinct antifungal activity exhibited by each strain. In addition to the still-to-be isolated and characterised unique compounds, these strains have been assessed for in planta production of alkaloids including lolitrem B, ergovaline, peramine, n-acetylloline, n-formylloline and epoxy janthitrems. These alkaloids are usually only considered in their role as mammalian or insect toxins. However, there is a possibility that these alkaloids contribute to the observed antifungal activity.
Only one study, in 1991, has tested Epichloë sp. derived alkaloids for their antifungal potential. Seigel and Latch used disk diffusion assays to evaluate n-formylloline, n-acetylloline, peramine sulfate, ergonovine maleate, ergotamine tartrate, and ergocryptine (100 mg/disk) for bioactivity against the phytopathogens Ceratobasidium cereale, Colletotrichum graminicola, Rhizoctonia zeae, and Limonomyces roseipellis, and in each instance they did not observe bioactivity despite the high concentrations of each compound used in the assay [18].
While the study revealed no significant antifungal activity from these compounds it does not account for three major alkaloids, ergovaline, lolitrem B and epoxy-janthitrem I or indole diterpene biosynthetic intermediates paxilline, terpendole E, and terpendole C. In the last decade, epoxy-janthitrems, a class of indole diterpenes that are structurally related to lolitrem B, were identified in Epichloë sp. endophyte strains including NEA12. Due to the inherent instability of pure epoxy janthitrems, stable isolation and purification remain elusive [32]. Consequently, there are limited reports on their sound biological activity in insects and mammals [33]. Since the study in 1991, other relevant purified compounds have also become available, thus providing an opportunity to screen and quantify a wider range of compounds for their bioprotective properties.
This study analyses the antifungal activity of thirteen Epichloë sp. related compounds against the phytopathogens Ceratobasidium sp., Fusarium sp., and Drechslera sp. to determine their contribution to the observed bioactivity previously described in Epichloë sp. endophyte strains by Fernando et al. (2020) [20].

Results
Bioprotective alkaloids known to be produced by Epichloë sp. strains utilised in pastures and turf were obtained from either commercial or in-house sources to assess their antifungal activity toward three phytopathogens Ceratobasidium sp., Drechslera sp., and Fusarium sp. (Table 1) in an agar well diffusion assay. Concentrations well above those observed in planta were prepared for each compound to provide the best prospects of detecting antifungal activity ( Table 2). The agar well diffusion assay results provide a visual representation of biological activity exhibited by these alkaloids. Antifungal activity is indicated by changes in the pathogen growth area, mycelial density, and direction of growth [20]. Image analysis using ImageJ software enables quantification of the activity by accurately measuring the pathogen growth area. The growth area measurements can then be used for statistical analysis.
Visual examination of agar well diffusion assays ( Figure 1) suggest there is no effect on the growth of the three pathogens Ceratobasidium sp., Drechslera sp., and Fusarium sp., with no changes in growth area, mycelial density, or direction of growth, in the presence of any of the tested alkaloids.
Visual observations from this study confirm previous findings by Siegel and Latch (1991) [18], and statistical analysis of quantified image data corroborates the findings. This study also further extends the observations made by Siegel and Latch (1991), to include Epichloë sp. (and related) alkaloids and pathway intermediates not previously investigated, such as ergovaline, ergocornine, ergotaminine, lolitrem B, janthitrem A, paxilline, terpendole E, and terpendole C.
The purity of the tested compounds, as described above, was greater than 95%. Each of the compounds used in this study have been shown to be in bioactive forms in other studies. For example, ergovaline, ergotamine, lolitrem B, paxilline, terpendole C, and terpendole E used in this study were previously used in investigations of mammalian toxicity in mice by Reddy et al. [6,36,37]. Paxilline and janthitrem A were used to investigate toxicity in mouse and insect models [32,38]. Ergot alkaloids ergotamine, ergocornine, ergocryptine, and ergotaminine have been widely used in investigations of mammalian toxicity [39]. Ergotamine, ergocornine, and ergocryptine are also known for their therapeutic application in medical conditions, e.g., migraine, uterotonics, cancer treatments, type II diabetes, and Parkinson's disease [40]. In instances where deuterated or salt derivatives of the compounds were used, the compounds may not display the same properties as in their natural forms. However, in the case of peramine hemisulfate, n-formylloline-d3, and n-acetylloline hydrochloride, the salts dissociate in solution to their natural forms, which are known for their insecticidal activity [41][42][43]. The salts and solvents were also pre-tested for their effects on pathogen growth in agar well diffusion assays, and 80% methanol, 16% HCl, and 1% lactic acid were included in the bioassay as negative controls.
The concentrations of the test alkaloids used in this study also provide an accurate indication of the likely occurrence of bioprotection conferred by individual compounds under natural conditions. To make an informed choice for the concentrations tested, average in planta concentrations were obtained from glasshouse-maintained perennial ryegrass-endophyte symbiota using liquid chromatography-mass spectrometry (LCMS) based detection and analysis of Epichloë sp. endophyte alkaloids [44]. It is important to note that alkaloid concentration varies enormously under different field conditions, such as season, temperature, and local environment [45,46], as well as during host plant establishment and development [47]. There is also significant host plant genotype-endophyte strain interaction that impacts quantitative alkaloid concentrations [45,46]. In this study, we analyzed average alkaloid abundance in planta for selected perennial ryegrass-endophyte symbiota, i.e., NEA12 (epoxy janthitrem I), NEA23 (lolines), and SE (peramine, lolitrem B, paxilline, terpendole E, terpendole C and ergovaline), grown under controlled glasshouse conditions. This was to inform the concentrations used in the bioassay as the abundance of some of these alkaloids produced in the field, particularly indole diterpene pathway intermediates and lolines in perennial ryegrass, has not been published. However, alkaloid abundances of well characterized endophyte-derived alkaloids, such as lolitrem B, ergova-line, and peramine, are higher in the glasshouse-maintained plants measured in this study than what is generally observed in the field [48,49].
While perennial ryegrass infected with Epichloë sp. endophytes provides protection against insect pests and diseases and thus is beneficial to pastoral agricultural systems, some of the major alkaloids are also detrimental to grazing animals at concentrations that result in toxicity (i.e., lolitrem B, paxilline, terpendole C, epoxy-janthitrem I, ergovaline). These alkaloids would not be suitable as biocontrol agents at toxic concentrations [5,6]. Hence, while it is possible that at higher concentrations the alkaloids would display some antifungal activity, they would be neither safe nor biologically relevant.
In a previous study by Fernando et al. (2020) [20], bioactive Epichloë sp. endophyte strains were identified in an in vitro screen for antifungal activity. A key observation was the differential bioactivity observed between three highly bioactive strains, namely NEA12, NEA23, and SE [20]. The three strains also exhibit different alkaloid profiles. NEA23 produces peramine, n-formylloline, and n-acetylloline, NEA12 produces epoxy janthitrems, and SE produces lolitrem B, peramine, and ergovaline [30,31,50]. This study establishes that the known, and well characterised in terms of animal and insect toxicity, endophyte-derived alkaloids are not responsible for the antifungal properties previously observed. It is now important to identify and characterise the compounds responsible for the antifungal activity of novel strains.

Conclusions
Epichloë sp. endophytes provide pasture and turf grasses with resistance to insect pests, animal herbivory, and disease resistance. For pastoral agriculture, the endophytes should maintain their insect and disease protection but contain little or none of the animal toxins that can cause disease in grazing animals. Although the compounds responsible for animal toxicity are largely known, the compounds responsible for disease resistance are not yet thoroughly described. In this study, the major alkaloids were tested for their in vitro antifungal activity. None of the tested compounds exhibited in vitro bioactivity towards the three phytopathogens tested. Since known alkaloid metabolites are not responsible for the antifungal activity expressed by Epichloë sp. endophyte strains, the bioactive metabolites are yet to be discovered. While this study provides an opportunity to rule out the contribution of alkaloids known to be produced by Epichloë endophyte strains, it opens avenues to explore novel metabolites produced by Epichloë endophytes.

Plant Material
All plant material was obtained from a glasshouse maintained (natural day lengths and a mean temperature of 22 • C) collection at Agriculture Victoria Research, Bundoora, Victoria, Australia.

Plant Pathogens
All pathogens (Table 1) were obtained from the National Collection of Fungi, Bundoora Herbarium, Victoria. Pathogens were stored as solid cultures on potato dextrose agar (PDA) plates (90 mm × 14 mm) (Sigma-Aldrich, Castle Hill, NSW, Australia) at 22 • C in the dark, and sub-cultured every two months to maintain stocks.
The genus of the pathogens used in this study was confirmed by internal transcribed spaces (ITS) sequence analysis described by Fernando et al. (2020) [20].

Agar Well Diffusion Assay
Bioassays were conducted on PDA plates (90 mm × 14 mm). A 4 mm cork borer was used to make two 4 mm diameter wells 2 cm apart. The bottoms of the wells were then sealed using 25 µL of PDA media to ensure the compound was retained closer to the surface. A 5 × 5 mm plug of pathogenic fungal mycelia was transferred onto the centre of the PDA media plate and each agar well was filled with 20 µL of the compound solution. The bioassay plates were prepared in replicates (n = 5). Plates were incubated at 22 • C in the dark for eight days and observations were taken daily from day three. Two negative controls and one positive control were prepared. For the negative controls, the agar wells were filled with either 20 µL of sterile distilled water or 4:1, v/v methanol: H 2 O or 1% lactic acid or 16% HCl. Positive controls were prepared by filling the agar wells with 20 µL of carbendazim [20]. Pathogen growth area (cm 2 ) was analysed using ImageJ 1× software (NIH, Bethesda, MA, USA). Image data of the sixth day of observations were used for the measurement of pathogen growth. One-way ANOVA was performed using Minitab ® 19 Statistical Software (Minitab, LLC, State College, PA, USA). Tukey post hoc comparison tables at 99% confidence level were used to determine significant differences in antifungal activity.